Task
Explore Lantern Documentation
As a developer looking to integrate AI functionalities, I explore the documentation for using Lantern effectively, so that I can understand how to generate embeddings and leverage its features in my applications.
Success definition: Given I have accessed the Lantern website When I click on the Docs link and navigate through the documentation sections Then I should successfully reach the 'Generate Embeddings' page to learn how to implement embedding generation.
Trajectory
Step 1:
Url (before/after):
https://lantern.dev/
https://lantern.dev/docs
Content (before/after):
RootWebArea Lantern, focused, url='https://lantern.dev/'
navigation
[57] link, center=(464,29), url='https://lantern.dev/'
image
[60] link Pricing, center=(786,29), url='https://lantern.dev/pricing'
Show more
[61] link Docs, center=(859,29), url='https://lantern.dev/docs'
[62] link Tutorials, center=(936,29), url='https://lantern.dev/tutorials'
[63] link Blog, center=(1012,29), url='https://lantern.dev/blog'
[64] link About, center=(1080,29), url='https://lantern.dev/about'
[65] link, center=(1144,29), url='https://github.com/lanterndata/lanterndb'
image
[68] link Login, center=(1406,29), url='https://lantern.dev/auth/sign-in'
[69] link Sign Up, center=(1506,29), url='https://lantern.dev/auth/sign-up'
heading Build in AI with Postgres
paragraph
image AI, url='https://lantern.dev/images/home/AI.png'
paragraph
paragraph
StaticText Lantern Cloud is an open-source Postgres vector database and toolkit for developers to build production-ready AI applications.
[86] link Get started for free, center=(852,354), url='https://lantern.dev/auth/sign-up'
paragraph
image
[89] link Explore the docs, center=(1094,354), url='https://lantern.dev/docs'
paragraph
StaticText Trusted by companies building for scale
image Owen Colegrove, url='https://storage.googleapis.com/lantern-web/OwenColegrove.jpg'
paragraph
StaticText Owen Colegrove
paragraph
StaticText Founder,
StaticText SciPhi
image SciPhi logo, url='https://storage.googleapis.com/lantern-web/SciphiLogo.png'
paragraph
StaticText Lantern helped us scale our cloud database infrastructure to support thousands of developers building RAG applications.
image
image Daksh Gupta, url='https://storage.googleapis.com/lantern-web/DakshGupta.jpg'
paragraph
StaticText Daksh Gupta
paragraph
StaticText Founder,
StaticText Greptile
image Greptile logo, url='https://storage.googleapis.com/lantern-web/GreptileLogo.png'
paragraph
StaticText We love Lantern! They made it easy for us to do search over thousands of repositories and millions of lines of code.
image
image Kevin Tang, url='https://storage.googleapis.com/lantern-web/KevinTang.jpeg'
paragraph
StaticText Kevin Tang
paragraph
StaticText Founder,
StaticText Firebender
image Firebender logo, url='https://storage.googleapis.com/lantern-web/FirebenderLogo.png'
paragraph
StaticText Lantern is the real deal; They helped fixed our Postgres setup and we were able to scale to 10M+ vectors with <50ms query times.
image
heading Vector search and BM25 in Postgres
paragraph
StaticText Get advanced search functionality inside Postgres with extensions -- no separate vector database or search engine required.
image
heading Vector search with pgvector
paragraph
StaticText Search over sparse and dense vectors, with support for binary, scalar, and product compression. Scale seamlessly to millions and beyond with serverless indexing.
StaticText BETA
image
heading Text search with BM25
paragraph
StaticText Use the BM25 ranking algorithm - the industry standard used in ElasticSearch - for more relevant text search results, surpassing the default full text search capabilities in Postgres.
StaticText BETA
image
heading Hybrid search
paragraph
StaticText Combine vector search and BM25 text search using RRF or other reranking algorithms for better results.
link Get started for free, url='https://lantern.dev/auth/sign-up'
paragraph
image
heading Scale to millions with serverless indexing
paragraph
StaticText Scale effortlessly without compromising database performance by offloading vector index creation to a separate machine from your main database.
image
list
listitem
image
paragraph
StaticText Infinite scalability
paragraph
StaticText Use as many or as little resources as you'd like, depending on your needs
listitem
image
paragraph
StaticText Performance
paragraph
StaticText Build or update indexes without slowing down your database
listitem
image
paragraph
StaticText Seamless integration
paragraph
StaticText Simply add external=true to vector index creation
link Get started for free, url='https://lantern.dev/auth/sign-up'
paragraph
image
heading Embedding generation and LLM integrations
paragraph
StaticText Simplify AI workflows by generating embeddings and running LLMs directly within your database.
list
listitem
image
paragraph
StaticText Generate vectors and run LLMs in Postgres
paragraph
StaticText Run simple SQL commands to generate vectors and run LLM models.
listitem
image
paragraph
StaticText Support for 20+ embedding models and LLMs
paragraph
StaticText Access Open AI, Cohere, Jina AI, and other open-source models.
listitem
image
paragraph
StaticText Automatically generate vector and LLM columns
paragraph
StaticText Create new vector or LLM columns based on your existing data.
link Get started for free, url='https://lantern.dev/auth/sign-up'
paragraph
image
StaticText Table: Products
paragraph
StaticText ID
paragraph
StaticText Name
image
paragraph
StaticText Description
image
paragraph
StaticText Vector
StaticText 1
StaticText Croissant
StaticText Flaky, buttery, soft breakfast pastry.
StaticText [0.2, 0.4, 0.7, 0.3, 0.8, 0.1, 0.5]
StaticText 2
StaticText Baguette
StaticText Long, thin loaf of French bread.
StaticText [0.1, 0.3, 0.6, 0.2, 0.7, 0.2, 0.4]
StaticText 3
StaticText Pain au chocolat
StaticText Chocolate-filled pastry.
StaticText Generating...
StaticText 4
StaticText Brioche
StaticText Generating...
StaticText Generating...
StaticText 5
StaticText Pretzel
StaticText Generating...
StaticText Generating...
StaticText 6
StaticText Sourdough
StaticText Generating...
StaticText Generating...
image
heading Easy to use — just use SQL or your favorite ORM
paragraph
StaticText No need to learn a new API or framework. Just use SQL or leverage our integrations with your favorite ORMs.
StaticText SQL
StaticText SQL
StaticText JAVASCRIPT
StaticText Sequelize
StaticText Knex
StaticText PYTHON
StaticText Django
link Docs, url='https://lantern.dev/docs/develop/get-started'
button Docs
image
code
StaticText 1
StaticText -- Create a table with a vector column
StaticText 2
StaticText CREATE
StaticText TABLE
StaticText books
StaticText (id
StaticText SERIAL
StaticText PRIMARY KEY
StaticText , book_embedding
StaticText REAL
StaticText [3]);
StaticText 3
StaticText 4
StaticText -- Insert some vectors
StaticText 5
StaticText INSERT INTO
StaticText books (book_embedding)
StaticText VALUES
StaticText (
StaticText '{0,1,0}'
StaticText ), (
StaticText '{3,2,4}'
StaticText );
StaticText 6
StaticText 7
StaticText -- Create an index for faster queries
StaticText 8
StaticText CREATE
StaticText INDEX
StaticText book_index
StaticText ON
StaticText books
StaticText USING
StaticText lantern_hnsw(book_embedding dist_l2sq_ops)
StaticText 9
StaticText WITH
StaticText (M
StaticText =
StaticText 2
StaticText , ef_construction
StaticText =
StaticText 10
StaticText , ef
StaticText =
StaticText 4
StaticText , dim
StaticText =
StaticText 3
StaticText );
StaticText 10
StaticText 11
StaticText -- Query the nearest vector
StaticText 12
StaticText SELECT
StaticText id
StaticText FROM
StaticText books
StaticText ORDER BY
StaticText book_embedding
StaticText <->
StaticText '{0,0,0}'
StaticText LIMIT
StaticText 1
StaticText ;
StaticText 13
StaticText 14
StaticText -- Query the nearest vector to a text embedding
StaticText 15
StaticText SELECT
StaticText id
StaticText FROM
StaticText books
StaticText ORDER BY
StaticText book_embedding
StaticText <->
StaticText text_embedding(
StaticText 'BAAI/bge-base-en'
StaticText ,
StaticText 'My text input'
StaticText )
StaticText LIMIT
StaticText 1
StaticText ;
link Get started for free, url='https://lantern.dev/auth/sign-up'
paragraph
image
link Explore the docs, url='https://lantern.dev/docs'
heading Lantern Cloud
paragraph
StaticText We're open-source, so you can self-host for full control, or use our managed service for easy setup and scaling.
paragraph
StaticText FREE FOREVER
heading Free tier
paragraph
StaticText A multi-tenant setup allows us to set up a free tier that never pauses.
paragraph
StaticText $0.00
StaticText /month
paragraph
StaticText All databases include
list
listitem
image
paragraph
StaticText Vector search and indexing
listitem
image
paragraph
StaticText Vector and LLM generation
listitem
image
paragraph
StaticText Automatic upgrades
listitem
image
paragraph
StaticText Query insights
listitem
image
paragraph
StaticText Community support
link Get started for free, url='https://lantern.dev/auth/sign-up'
paragraph
StaticText BEST CHOICE
heading Production
paragraph
StaticText Designed for running high-performance production workloads
paragraph
StaticText STARTING AT
paragraph
StaticText $44.00
StaticText /month
paragraph
StaticText Everything in free, plus
list
listitem
image
paragraph
StaticText Increased compute and storage
listitem
image
paragraph
StaticText Automatic backups
listitem
image
paragraph
StaticText Point-in-time recovery
listitem
image
paragraph
StaticText Asynchronous tasks
listitem
image
paragraph
StaticText Priority support
link Get started, url='https://lantern.dev/auth/sign-up'
heading Start building
link Get started for free, url='https://lantern.dev/auth/sign-up'
paragraph
image
link Explore the docs, url='https://lantern.dev/docs'
paragraph
StaticText Developers
link Documentation, url='https://lantern.dev/docs'
paragraph
link API Reference, url='https://lantern.dev/docs'
paragraph
link Tutorials, url='https://lantern.dev/Examples'
paragraph
link Blog, url='https://lantern.dev/blog'
paragraph
paragraph
StaticText Company
link About, url='https://lantern.dev/about'
paragraph
link Pricing, url='https://lantern.dev/pricing'
paragraph
link Privacy Policy, url='https://lantern.dev/privacy'
paragraph
link Terms & Conditions, url='https://lantern.dev/terms'
paragraph
paragraph
StaticText Contact
link Github, url='https://github.com/lanterndata/lantern'
paragraph
link Twitter, url='https://twitter.com/lanterndb'
paragraph
link LinkedIn, url='https://www.linkedin.com/company/98218505'
paragraph
link Email, url='mailto:support@lantern.dev'
paragraph
link, url='https://lantern.dev/'
image
paragraph
StaticText © 2024 Lantern Systems, Inc.
alert, atomic
RootWebArea Docs | Lantern, focused, url='https://lantern.dev/docs'
alert, atomic
StaticText Docs | Lantern
navigation
[567] link, center=(128,29), url='https://lantern.dev/'
Show more
image
[570] link Pricing, center=(786,29), url='https://lantern.dev/pricing'
[571] link Docs, center=(859,29), url='https://lantern.dev/docs'
[572] link Tutorials, center=(936,29), url='https://lantern.dev/tutorials'
[573] link Blog, center=(1012,29), url='https://lantern.dev/blog'
[574] link About, center=(1080,29), url='https://lantern.dev/about'
[575] link, center=(1144,29), url='https://github.com/lanterndata/lanterndb'
image
[578] link Login, center=(1742,29), url='https://lantern.dev/auth/sign-in'
[579] link Sign Up, center=(1842,29), url='https://lantern.dev/auth/sign-up'
main
navigation
[590] button Search K, center=(144,89), inner_text=Search
K
image
paragraph
image
paragraph
StaticText Develop
[598] link Get Started, center=(148,192), url='https://lantern.dev/docs/develop/get-started'
[600] link Store Embeddings, center=(148,224), url='https://lantern.dev/docs/develop/store'
[602] link Generate Embeddings, center=(148,256), url='https://lantern.dev/docs/develop/generate'
[604] link Automatic Embedding Generation, center=(148,298), url='https://lantern.dev/docs/develop/columns'
[606] link Calculate Distance, center=(148,340), url='https://lantern.dev/docs/develop/distance'
[608] link Query Embeddings, center=(148,372), url='https://lantern.dev/docs/develop/query'
[610] link Create Index, center=(148,404), url='https://lantern.dev/docs/develop/indexing'
[612] link Quantization, center=(148,436), url='https://lantern.dev/docs/develop/quantization'
[614] link Asynchronous Tasks, center=(148,468), url='https://lantern.dev/docs/develop/async-tasks'
[616] link Weighted Vector Search, center=(148,500), url='https://lantern.dev/docs/develop/weighted-vector-search'
[618] link Troubleshooting, center=(148,532), url='https://lantern.dev/docs/develop/troubleshooting'
[620] link Single Operator, center=(148,564), url='https://lantern.dev/docs/develop/operator'
[622] link Postgres Notes, center=(148,596), url='https://lantern.dev/docs/develop/postgres'
[624] link Security, center=(148,628), url='https://lantern.dev/docs/develop/security'
paragraph
StaticText Languages
[628] link Javascript, center=(148,712), url='https://lantern.dev/docs/languages/javascript'
[630] link Python, center=(148,744), url='https://lantern.dev/docs/languages/python'
[632] link Ruby, center=(148,776), url='https://lantern.dev/docs/languages/ruby'
[634] link Rust, center=(148,808), url='https://lantern.dev/docs/languages/rust'
paragraph
StaticText Migrate
[638] link Migrate from Postgres to Lantern Cloud, center=(148,902), url='https://lantern.dev/docs/migrate/postgres-cloud'
[640] link Migrate from pgvector to Lantern Cloud, center=(148,954), url='https://lantern.dev/docs/migrate/pgvector-cloud'
[642] link Migrate from pgvector to self-hosted Lantern, center=(148,1006), url='https://lantern.dev/docs/migrate/pgvector-self'
[644] link Migrate from Pinecone to Lantern Cloud, center=(148,1058), url='https://lantern.dev/docs/migrate/pinecone'
paragraph
StaticText Lantern HNSW
link Installation, url='https://lantern.dev/docs/lantern-hnsw/install'
paragraph
StaticText Lantern Extras
link Installation, url='https://lantern.dev/docs/lantern-extras/install'
link Generate Embeddings, url='https://lantern.dev/docs/lantern-extras/embeddings'
link Lantern Daemon, url='https://lantern.dev/docs/lantern-extras/daemon'
paragraph
StaticText Lantern CLI
link Installation, url='https://lantern.dev/docs/lantern-cli/install'
link Generate Embeddings, url='https://lantern.dev/docs/lantern-cli/embeddings'
link Indexing Server, url='https://lantern.dev/docs/lantern-cli/indexing'
link Daemon, url='https://lantern.dev/docs/lantern-cli/daemon'
link Autotune Index, url='https://lantern.dev/docs/lantern-cli/autotune'
link Product Quantization, url='https://lantern.dev/docs/lantern-cli/pq'
paragraph
StaticText Contributing
link Dockerfile, url='https://lantern.dev/docs/contributing/docker'
link Visual Studio Code, url='https://lantern.dev/docs/contributing/vscode'
heading Build AI Applications with Postgres
paragraph
StaticText Lantern is the only database you need for your AI applications.
list
[683] listitem, center=(1104,204), inner_text=Built on top of Postgres- Add AI to your existing applications by just writing another SQL query. We support Postgres 11 - 16.
ListMarker •
StaticText Built on top of Postgres
StaticText -
StaticText Add AI to your existing applications by just writing another SQL query. We support Postgres 11 - 16.
listitem
ListMarker •
StaticText Fast index creation that scales to billions of vectors
StaticText -
StaticText Create an HNSW index for your data in minutes (see our
[689] link benchmarks, center=(1372,236), url='https://lantern.dev/blog/hnsw-index-creation'
StaticText ).
[690] listitem, center=(1104,280), inner_text=Create embeddings without worrying about infrastructure- Simply select the model you'd like to use to embed your data, and we'll handle the rest. No need to worry about finding the right library, switching languages, or handling API call failures.
ListMarker •
StaticText Create embeddings without worrying about infrastructure
StaticText -
StaticText Simply select the model you'd like to use to embed your data, and we'll handle the rest. No need to worry about finding the right library, switching languages, or handling API call failures.
[693] listitem, center=(1104,336), inner_text=Index hyperparameter optimization- We'll automatically tune your index for you, so you don't have to worry about finding the right set of parameters, and changing them over time.
ListMarker •
StaticText Index hyperparameter optimization
StaticText -
StaticText We'll automatically tune your index for you, so you don't have to worry about finding the right set of parameters, and changing them over time.
heading Getting Started
paragraph
StaticText The easiest way to get started with Lantern is with
[698] link Lantern Cloud, center=(908,456), url='https://lantern.dev/'
StaticText . Lantern Cloud is a fully managed database offering with support for embedding generation and management. We also provide
[699] link guides, center=(744,480), url='https://lantern.dev/docs/develop/get-started'
StaticText on self-hosting Lantern.
[700] link Sign up for Lantern Cloud, center=(626,538), url='https://lantern.dev/auth/sign-up'
heading Explore Documentation
paragraph
StaticText Check our
[703] link development guide, center=(650,660), url='https://lantern.dev/docs/develop/store'
StaticText to learn more about how to use Lantern. We support generating vectors with a a variety of embedding models out of the box. With a single operator, you can query your vectors using a range of distance functions, and filter these results using your existing data.
[705] link Get Started, center=(800,746), url='https://lantern.dev/docs/develop/get-started'
image
paragraph
[710] link Store Embeddings, center=(1408,746), url='https://lantern.dev/docs/develop/store'
image
paragraph
[715] link Create Index, center=(800,824), url='https://lantern.dev/docs/develop/index'
image
paragraph
[720] link Query Embeddings, center=(1408,824), url='https://lantern.dev/docs/develop/query'
image
paragraph
heading Quickstart Tutorials
paragraph
StaticText Integrate Lantern and vector search into an existing application by following one of our quickstart tutorials.
[728] link Python, center=(800,1014), url='https://lantern.dev/tutorials/python/quickstart'
image
paragraph
[733] link Javascript, center=(1408,1014), url='https://lantern.dev/tutorials/javascript/quickstart'
image
paragraph
link Java, url='https://lantern.dev/tutorials/java/quickstart'
image
paragraph
link Ruby, url='https://lantern.dev/tutorials/ruby/quickstart'
image
paragraph
heading Get Help
paragraph
StaticText If you need any help with using Lantern, reach out to our support team at
link support@lantern.dev., url='mailto:support@lantern.dev'
Reasoning for action: By clicking on the Docs link, I successfully transition from the homepage to the documentation section, which is essential for further exploration of how to generate embeddings and leverage the features of the Lantern platform. This is the first necessary step towards reaching the specific 'Generate Embeddings' page outlined in the user story.
Action: click(61) # Click on the Docs link to explore the documentation section.
Change Description: The browser navigated from the homepage to the documentation section of the Lantern website, updating the URL to 'https://lantern.dev/docs' and changing the focused content to display the documentation links and information, while retaining the main navigation options at the top.
Screenshot before:Screenshot after:---
Step 2:
Url (before/after):
https://lantern.dev/docs
https://lantern.dev/docs/develop/get-started
Content (before/after):
RootWebArea Docs | Lantern, focused, url='https://lantern.dev/docs'
alert, atomic
StaticText Docs | Lantern
navigation
[567] link, center=(128,29), url='https://lantern.dev/'
Show more
image
[570] link Pricing, center=(786,29), url='https://lantern.dev/pricing'
[571] link Docs, center=(859,29), url='https://lantern.dev/docs'
[572] link Tutorials, center=(936,29), url='https://lantern.dev/tutorials'
[573] link Blog, center=(1012,29), url='https://lantern.dev/blog'
[574] link About, center=(1080,29), url='https://lantern.dev/about'
[575] link, center=(1144,29), url='https://github.com/lanterndata/lanterndb'
image
[578] link Login, center=(1742,29), url='https://lantern.dev/auth/sign-in'
[579] link Sign Up, center=(1842,29), url='https://lantern.dev/auth/sign-up'
main
navigation
[590] button Search K, center=(144,89), inner_text=Search
K
image
paragraph
image
paragraph
StaticText Develop
[598] link Get Started, center=(148,192), url='https://lantern.dev/docs/develop/get-started'
[600] link Store Embeddings, center=(148,224), url='https://lantern.dev/docs/develop/store'
[602] link Generate Embeddings, center=(148,256), url='https://lantern.dev/docs/develop/generate'
[604] link Automatic Embedding Generation, center=(148,298), url='https://lantern.dev/docs/develop/columns'
[606] link Calculate Distance, center=(148,340), url='https://lantern.dev/docs/develop/distance'
[608] link Query Embeddings, center=(148,372), url='https://lantern.dev/docs/develop/query'
[610] link Create Index, center=(148,404), url='https://lantern.dev/docs/develop/indexing'
[612] link Quantization, center=(148,436), url='https://lantern.dev/docs/develop/quantization'
[614] link Asynchronous Tasks, center=(148,468), url='https://lantern.dev/docs/develop/async-tasks'
[616] link Weighted Vector Search, center=(148,500), url='https://lantern.dev/docs/develop/weighted-vector-search'
[618] link Troubleshooting, center=(148,532), url='https://lantern.dev/docs/develop/troubleshooting'
[620] link Single Operator, center=(148,564), url='https://lantern.dev/docs/develop/operator'
[622] link Postgres Notes, center=(148,596), url='https://lantern.dev/docs/develop/postgres'
[624] link Security, center=(148,628), url='https://lantern.dev/docs/develop/security'
paragraph
StaticText Languages
[628] link Javascript, center=(148,712), url='https://lantern.dev/docs/languages/javascript'
[630] link Python, center=(148,744), url='https://lantern.dev/docs/languages/python'
[632] link Ruby, center=(148,776), url='https://lantern.dev/docs/languages/ruby'
[634] link Rust, center=(148,808), url='https://lantern.dev/docs/languages/rust'
paragraph
StaticText Migrate
[638] link Migrate from Postgres to Lantern Cloud, center=(148,902), url='https://lantern.dev/docs/migrate/postgres-cloud'
[640] link Migrate from pgvector to Lantern Cloud, center=(148,954), url='https://lantern.dev/docs/migrate/pgvector-cloud'
[642] link Migrate from pgvector to self-hosted Lantern, center=(148,1006), url='https://lantern.dev/docs/migrate/pgvector-self'
[644] link Migrate from Pinecone to Lantern Cloud, center=(148,1058), url='https://lantern.dev/docs/migrate/pinecone'
paragraph
StaticText Lantern HNSW
link Installation, url='https://lantern.dev/docs/lantern-hnsw/install'
paragraph
StaticText Lantern Extras
link Installation, url='https://lantern.dev/docs/lantern-extras/install'
link Generate Embeddings, url='https://lantern.dev/docs/lantern-extras/embeddings'
link Lantern Daemon, url='https://lantern.dev/docs/lantern-extras/daemon'
paragraph
StaticText Lantern CLI
link Installation, url='https://lantern.dev/docs/lantern-cli/install'
link Generate Embeddings, url='https://lantern.dev/docs/lantern-cli/embeddings'
link Indexing Server, url='https://lantern.dev/docs/lantern-cli/indexing'
link Daemon, url='https://lantern.dev/docs/lantern-cli/daemon'
link Autotune Index, url='https://lantern.dev/docs/lantern-cli/autotune'
link Product Quantization, url='https://lantern.dev/docs/lantern-cli/pq'
paragraph
StaticText Contributing
link Dockerfile, url='https://lantern.dev/docs/contributing/docker'
link Visual Studio Code, url='https://lantern.dev/docs/contributing/vscode'
heading Build AI Applications with Postgres
paragraph
StaticText Lantern is the only database you need for your AI applications.
list
[683] listitem, center=(1104,204), inner_text=Built on top of Postgres- Add AI to your existing applications by just writing another SQL query. We support Postgres 11 - 16.
ListMarker •
StaticText Built on top of Postgres
StaticText -
StaticText Add AI to your existing applications by just writing another SQL query. We support Postgres 11 - 16.
listitem
ListMarker •
StaticText Fast index creation that scales to billions of vectors
StaticText -
StaticText Create an HNSW index for your data in minutes (see our
[689] link benchmarks, center=(1372,236), url='https://lantern.dev/blog/hnsw-index-creation'
StaticText ).
[690] listitem, center=(1104,280), inner_text=Create embeddings without worrying about infrastructure- Simply select the model you'd like to use to embed your data, and we'll handle the rest. No need to worry about finding the right library, switching languages, or handling API call failures.
ListMarker •
StaticText Create embeddings without worrying about infrastructure
StaticText -
StaticText Simply select the model you'd like to use to embed your data, and we'll handle the rest. No need to worry about finding the right library, switching languages, or handling API call failures.
[693] listitem, center=(1104,336), inner_text=Index hyperparameter optimization- We'll automatically tune your index for you, so you don't have to worry about finding the right set of parameters, and changing them over time.
ListMarker •
StaticText Index hyperparameter optimization
StaticText -
StaticText We'll automatically tune your index for you, so you don't have to worry about finding the right set of parameters, and changing them over time.
heading Getting Started
paragraph
StaticText The easiest way to get started with Lantern is with
[698] link Lantern Cloud, center=(908,456), url='https://lantern.dev/'
StaticText . Lantern Cloud is a fully managed database offering with support for embedding generation and management. We also provide
[699] link guides, center=(744,480), url='https://lantern.dev/docs/develop/get-started'
StaticText on self-hosting Lantern.
[700] link Sign up for Lantern Cloud, center=(626,538), url='https://lantern.dev/auth/sign-up'
heading Explore Documentation
paragraph
StaticText Check our
[703] link development guide, center=(650,660), url='https://lantern.dev/docs/develop/store'
StaticText to learn more about how to use Lantern. We support generating vectors with a a variety of embedding models out of the box. With a single operator, you can query your vectors using a range of distance functions, and filter these results using your existing data.
[705] link Get Started, center=(800,746), url='https://lantern.dev/docs/develop/get-started'
image
paragraph
[710] link Store Embeddings, center=(1408,746), url='https://lantern.dev/docs/develop/store'
image
paragraph
[715] link Create Index, center=(800,824), url='https://lantern.dev/docs/develop/index'
image
paragraph
[720] link Query Embeddings, center=(1408,824), url='https://lantern.dev/docs/develop/query'
image
paragraph
heading Quickstart Tutorials
paragraph
StaticText Integrate Lantern and vector search into an existing application by following one of our quickstart tutorials.
[728] link Python, center=(800,1014), url='https://lantern.dev/tutorials/python/quickstart'
image
paragraph
[733] link Javascript, center=(1408,1014), url='https://lantern.dev/tutorials/javascript/quickstart'
image
paragraph
link Java, url='https://lantern.dev/tutorials/java/quickstart'
image
paragraph
link Ruby, url='https://lantern.dev/tutorials/ruby/quickstart'
image
paragraph
heading Get Help
paragraph
StaticText If you need any help with using Lantern, reach out to our support team at
link support@lantern.dev., url='mailto:support@lantern.dev'
RootWebArea Develop - Get Started | Lantern Docs, focused, url='https://lantern.dev/docs/develop/get-started'
alert, atomic
StaticText Develop - Get Started | Lantern Docs
navigation
[567] link, center=(128,29), url='https://lantern.dev/'
Show more
image
[570] link Pricing, center=(786,29), url='https://lantern.dev/pricing'
[571] link Docs, center=(859,29), url='https://lantern.dev/docs'
[572] link Tutorials, center=(936,29), url='https://lantern.dev/tutorials'
[573] link Blog, center=(1012,29), url='https://lantern.dev/blog'
[574] link About, center=(1080,29), url='https://lantern.dev/about'
[575] link, center=(1144,29), url='https://github.com/lanterndata/lanterndb'
image
[578] link Login, center=(1742,29), url='https://lantern.dev/auth/sign-in'
[579] link Sign Up, center=(1842,29), url='https://lantern.dev/auth/sign-up'
main
navigation
[590] button Search K, center=(144,89), inner_text=Search
K
image
paragraph
image
paragraph
StaticText Develop
[598] link Get Started, center=(148,192), focused, url='https://lantern.dev/docs/develop/get-started'
[600] link Store Embeddings, center=(148,224), url='https://lantern.dev/docs/develop/store'
[602] link Generate Embeddings, center=(148,256), url='https://lantern.dev/docs/develop/generate'
[604] link Automatic Embedding Generation, center=(148,298), url='https://lantern.dev/docs/develop/columns'
[606] link Calculate Distance, center=(148,340), url='https://lantern.dev/docs/develop/distance'
[608] link Query Embeddings, center=(148,372), url='https://lantern.dev/docs/develop/query'
[610] link Create Index, center=(148,404), url='https://lantern.dev/docs/develop/indexing'
[612] link Quantization, center=(148,436), url='https://lantern.dev/docs/develop/quantization'
[614] link Asynchronous Tasks, center=(148,468), url='https://lantern.dev/docs/develop/async-tasks'
[616] link Weighted Vector Search, center=(148,500), url='https://lantern.dev/docs/develop/weighted-vector-search'
[618] link Troubleshooting, center=(148,532), url='https://lantern.dev/docs/develop/troubleshooting'
[620] link Single Operator, center=(148,564), url='https://lantern.dev/docs/develop/operator'
[622] link Postgres Notes, center=(148,596), url='https://lantern.dev/docs/develop/postgres'
[624] link Security, center=(148,628), url='https://lantern.dev/docs/develop/security'
paragraph
StaticText Languages
[628] link Javascript, center=(148,712), url='https://lantern.dev/docs/languages/javascript'
[630] link Python, center=(148,744), url='https://lantern.dev/docs/languages/python'
[632] link Ruby, center=(148,776), url='https://lantern.dev/docs/languages/ruby'
[634] link Rust, center=(148,808), url='https://lantern.dev/docs/languages/rust'
paragraph
StaticText Migrate
[638] link Migrate from Postgres to Lantern Cloud, center=(148,902), url='https://lantern.dev/docs/migrate/postgres-cloud'
[640] link Migrate from pgvector to Lantern Cloud, center=(148,954), url='https://lantern.dev/docs/migrate/pgvector-cloud'
[642] link Migrate from pgvector to self-hosted Lantern, center=(148,1006), url='https://lantern.dev/docs/migrate/pgvector-self'
[644] link Migrate from Pinecone to Lantern Cloud, center=(148,1058), url='https://lantern.dev/docs/migrate/pinecone'
paragraph
StaticText Lantern HNSW
link Installation, url='https://lantern.dev/docs/lantern-hnsw/install'
paragraph
StaticText Lantern Extras
link Installation, url='https://lantern.dev/docs/lantern-extras/install'
link Generate Embeddings, url='https://lantern.dev/docs/lantern-extras/embeddings'
link Lantern Daemon, url='https://lantern.dev/docs/lantern-extras/daemon'
paragraph
StaticText Lantern CLI
link Installation, url='https://lantern.dev/docs/lantern-cli/install'
link Generate Embeddings, url='https://lantern.dev/docs/lantern-cli/embeddings'
link Indexing Server, url='https://lantern.dev/docs/lantern-cli/indexing'
link Daemon, url='https://lantern.dev/docs/lantern-cli/daemon'
link Autotune Index, url='https://lantern.dev/docs/lantern-cli/autotune'
link Product Quantization, url='https://lantern.dev/docs/lantern-cli/pq'
paragraph
StaticText Contributing
link Dockerfile, url='https://lantern.dev/docs/contributing/docker'
link Visual Studio Code, url='https://lantern.dev/docs/contributing/vscode'
article
paragraph
StaticText Develop
heading Get Started
heading Lantern Cloud
paragraph
StaticText The easiest way to get started with all of our tools is with
[779] link Lantern Cloud, center=(905,268), url='https://lantern.dev/'
StaticText .
paragraph
StaticText In Lantern Cloud, you can create create a database with just a few clicks. Once you load your data into Lantern, you can
link generate embeddings, url='https://lantern.dev/docs/develop/generate'
StaticText with a single click from dozens of provided open source or proprietary embedding models. You can then
[782] link create a vector index, center=(1364,336), url='https://lantern.dev/docs/develop/indexing'
StaticText from your dashboard or run an index-tuning experiment to choose the best parameters for index creation.
heading Ubicloud
paragraph
StaticText Lantern is also available as a managed offering on
[785] link Ubicloud, center=(838,484), url='https://ubicloud.com/'
StaticText . You'll still receive direct support from the Lantern team for instances on Ubicloud.
paragraph
StaticText Ubicloud is an open-source alternative to AWS. It offers cloud services like virtual machines, block storage, and managed Postgres at rates 2-10x lower than hyperscalers like AWS, Azure, and GCP.
heading Self-Host
paragraph
StaticText Alternatively, you can also use our tools locally or self-host them. There are three tools that are provided out-of-the-box with Lantern Cloud.
list
listitem
ListMarker •
[791] link Lantern HNSW, center=(534,720), url='https://lantern.dev/docs/lantern-hnsw/install'
StaticText , our core Postgres extension, provides vector search in Postgres.
listitem
ListMarker •
[793] link Lantern Extras, center=(532,760), url='https://lantern.dev/docs/lantern-extras/install'
StaticText , which further extends Postgres to support embedding generation.
listitem
ListMarker •
[795] link Lantern CLI, center=(522,800), url='https://lantern.dev/docs/lantern-cli/install'
StaticText provides routines for generating embeddings and indexes.
paragraph
StaticText You can install the tools individually by following the instructions linked.
heading Overview
paragraph
StaticText Here is a non-comprehensive overview of what you can do with Lantern. The examples below use SQL, but we also provide examples for
[799] link Python, center=(482,992), url='https://lantern.dev/docs/languages/python'
StaticText ,
[800] link JavaScript, center=(554,992), url='https://lantern.dev/docs/languages/javascript'
StaticText ,
[801] link Ruby, center=(620,992), url='https://lantern.dev/docs/languages/ruby'
StaticText , and
[802] link Rust, center=(693,992), url='https://lantern.dev/docs/languages/rust'
StaticText .
paragraph
StaticText Create a table with an embedding column
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText CREATE
StaticText TABLE
StaticText books
StaticText (id
StaticText SERIAL
StaticText PRIMARY KEY
StaticText , book_embedding
StaticText REAL
StaticText [3]);
paragraph
StaticText Generate embeddings
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText SELECT
StaticText text_embedding(
StaticText 'BAAI/bge-base-en'
StaticText ,
StaticText 'My text input'
StaticText );
StaticText 2
StaticText SELECT
StaticText openai_embedding(
StaticText 'openai/text-embedding-ada-002'
StaticText ,
StaticText 'My text input'
StaticText );
paragraph
StaticText Insert embeddings
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText INSERT INTO
StaticText books (book_embedding)
StaticText VALUES
StaticText (
StaticText '{0,1,0}'
StaticText ), (
StaticText '{3,2,4}'
StaticText );
paragraph
StaticText Calculate distance and select nearest rows
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText SELECT
StaticText book_embedding
StaticText <->
StaticText '{0,0,0}'
StaticText FROM
StaticText books
StaticText 2
StaticText ORDER BY
StaticText book_embedding
StaticText <->
StaticText '{0,0,0}'
StaticText LIMIT
StaticText 1
StaticText ;
paragraph
StaticText Create an index to speed up nearest neighbor search
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText CREATE
StaticText INDEX
StaticText book_index
StaticText ON
StaticText books
StaticText USING
StaticText lantern_hnsw(book_embedding dist_l2sq_ops)
StaticText 2
StaticText WITH
StaticText (M
StaticText =
StaticText 2
StaticText , ef_construction
StaticText =
StaticText 10
StaticText , ef
StaticText =
StaticText 4
StaticText , dim
StaticText =
StaticText 3
StaticText );
link Edit this page, url='https://github.com/lanterndata/docs/blob/main/010-develop/000-get-started.md'
paragraph
paragraph
StaticText ON THIS PAGE
list
[931] listitem, center=(1776,122), inner_text=Lantern Cloud
StaticText Lantern Cloud
[932] listitem, center=(1776,158), inner_text=Ubicloud
StaticText Ubicloud
[933] listitem, center=(1776,194), inner_text=Self-Host
StaticText Self-Host
[934] listitem, center=(1776,230), inner_text=Overview
StaticText Overview
Reasoning for action: I initiated the action of clicking the 'Get Started' link because this page typically contains essential introductory content, tutorials, and details necessary for new users to effectively use Lantern. This step is important as it can provide context and lead me towards further links or sections including the 'Generate Embeddings' page, which is the ultimate goal of the user story.
Action: click(598) # Click on the 'Get Started' link to explore foundational information about using Lantern.
Change Description: The user clicked on the 'Get Started' link, resulting in a navigation change to the 'Develop - Get Started' page of the Lantern Docs, with the URL updated to 'https://lantern.dev/docs/develop/get-started', and the focus shifted to the content specific to getting started with Lantern.
Screenshot before:Screenshot after:---
Step 3:
Url (before/after):
https://lantern.dev/docs/develop/get-started
https://lantern.dev/docs/develop/generate
Content (before/after):
RootWebArea Develop - Get Started | Lantern Docs, focused, url='https://lantern.dev/docs/develop/get-started'
alert, atomic
StaticText Develop - Get Started | Lantern Docs
navigation
[567] link, center=(128,29), url='https://lantern.dev/'
Show more
image
[570] link Pricing, center=(786,29), url='https://lantern.dev/pricing'
[571] link Docs, center=(859,29), url='https://lantern.dev/docs'
[572] link Tutorials, center=(936,29), url='https://lantern.dev/tutorials'
[573] link Blog, center=(1012,29), url='https://lantern.dev/blog'
[574] link About, center=(1080,29), url='https://lantern.dev/about'
[575] link, center=(1144,29), url='https://github.com/lanterndata/lanterndb'
image
[578] link Login, center=(1742,29), url='https://lantern.dev/auth/sign-in'
[579] link Sign Up, center=(1842,29), url='https://lantern.dev/auth/sign-up'
main
navigation
[590] button Search K, center=(144,89), inner_text=Search
K
image
paragraph
image
paragraph
StaticText Develop
[598] link Get Started, center=(148,192), focused, url='https://lantern.dev/docs/develop/get-started'
[600] link Store Embeddings, center=(148,224), url='https://lantern.dev/docs/develop/store'
[602] link Generate Embeddings, center=(148,256), url='https://lantern.dev/docs/develop/generate'
[604] link Automatic Embedding Generation, center=(148,298), url='https://lantern.dev/docs/develop/columns'
[606] link Calculate Distance, center=(148,340), url='https://lantern.dev/docs/develop/distance'
[608] link Query Embeddings, center=(148,372), url='https://lantern.dev/docs/develop/query'
[610] link Create Index, center=(148,404), url='https://lantern.dev/docs/develop/indexing'
[612] link Quantization, center=(148,436), url='https://lantern.dev/docs/develop/quantization'
[614] link Asynchronous Tasks, center=(148,468), url='https://lantern.dev/docs/develop/async-tasks'
[616] link Weighted Vector Search, center=(148,500), url='https://lantern.dev/docs/develop/weighted-vector-search'
[618] link Troubleshooting, center=(148,532), url='https://lantern.dev/docs/develop/troubleshooting'
[620] link Single Operator, center=(148,564), url='https://lantern.dev/docs/develop/operator'
[622] link Postgres Notes, center=(148,596), url='https://lantern.dev/docs/develop/postgres'
[624] link Security, center=(148,628), url='https://lantern.dev/docs/develop/security'
paragraph
StaticText Languages
[628] link Javascript, center=(148,712), url='https://lantern.dev/docs/languages/javascript'
[630] link Python, center=(148,744), url='https://lantern.dev/docs/languages/python'
[632] link Ruby, center=(148,776), url='https://lantern.dev/docs/languages/ruby'
[634] link Rust, center=(148,808), url='https://lantern.dev/docs/languages/rust'
paragraph
StaticText Migrate
[638] link Migrate from Postgres to Lantern Cloud, center=(148,902), url='https://lantern.dev/docs/migrate/postgres-cloud'
[640] link Migrate from pgvector to Lantern Cloud, center=(148,954), url='https://lantern.dev/docs/migrate/pgvector-cloud'
[642] link Migrate from pgvector to self-hosted Lantern, center=(148,1006), url='https://lantern.dev/docs/migrate/pgvector-self'
[644] link Migrate from Pinecone to Lantern Cloud, center=(148,1058), url='https://lantern.dev/docs/migrate/pinecone'
paragraph
StaticText Lantern HNSW
link Installation, url='https://lantern.dev/docs/lantern-hnsw/install'
paragraph
StaticText Lantern Extras
link Installation, url='https://lantern.dev/docs/lantern-extras/install'
link Generate Embeddings, url='https://lantern.dev/docs/lantern-extras/embeddings'
link Lantern Daemon, url='https://lantern.dev/docs/lantern-extras/daemon'
paragraph
StaticText Lantern CLI
link Installation, url='https://lantern.dev/docs/lantern-cli/install'
link Generate Embeddings, url='https://lantern.dev/docs/lantern-cli/embeddings'
link Indexing Server, url='https://lantern.dev/docs/lantern-cli/indexing'
link Daemon, url='https://lantern.dev/docs/lantern-cli/daemon'
link Autotune Index, url='https://lantern.dev/docs/lantern-cli/autotune'
link Product Quantization, url='https://lantern.dev/docs/lantern-cli/pq'
paragraph
StaticText Contributing
link Dockerfile, url='https://lantern.dev/docs/contributing/docker'
link Visual Studio Code, url='https://lantern.dev/docs/contributing/vscode'
article
paragraph
StaticText Develop
heading Get Started
heading Lantern Cloud
paragraph
StaticText The easiest way to get started with all of our tools is with
[779] link Lantern Cloud, center=(905,268), url='https://lantern.dev/'
StaticText .
paragraph
StaticText In Lantern Cloud, you can create create a database with just a few clicks. Once you load your data into Lantern, you can
link generate embeddings, url='https://lantern.dev/docs/develop/generate'
StaticText with a single click from dozens of provided open source or proprietary embedding models. You can then
[782] link create a vector index, center=(1364,336), url='https://lantern.dev/docs/develop/indexing'
StaticText from your dashboard or run an index-tuning experiment to choose the best parameters for index creation.
heading Ubicloud
paragraph
StaticText Lantern is also available as a managed offering on
[785] link Ubicloud, center=(838,484), url='https://ubicloud.com/'
StaticText . You'll still receive direct support from the Lantern team for instances on Ubicloud.
paragraph
StaticText Ubicloud is an open-source alternative to AWS. It offers cloud services like virtual machines, block storage, and managed Postgres at rates 2-10x lower than hyperscalers like AWS, Azure, and GCP.
heading Self-Host
paragraph
StaticText Alternatively, you can also use our tools locally or self-host them. There are three tools that are provided out-of-the-box with Lantern Cloud.
list
listitem
ListMarker •
[791] link Lantern HNSW, center=(534,720), url='https://lantern.dev/docs/lantern-hnsw/install'
StaticText , our core Postgres extension, provides vector search in Postgres.
listitem
ListMarker •
[793] link Lantern Extras, center=(532,760), url='https://lantern.dev/docs/lantern-extras/install'
StaticText , which further extends Postgres to support embedding generation.
listitem
ListMarker •
[795] link Lantern CLI, center=(522,800), url='https://lantern.dev/docs/lantern-cli/install'
StaticText provides routines for generating embeddings and indexes.
paragraph
StaticText You can install the tools individually by following the instructions linked.
heading Overview
paragraph
StaticText Here is a non-comprehensive overview of what you can do with Lantern. The examples below use SQL, but we also provide examples for
[799] link Python, center=(482,992), url='https://lantern.dev/docs/languages/python'
StaticText ,
[800] link JavaScript, center=(554,992), url='https://lantern.dev/docs/languages/javascript'
StaticText ,
[801] link Ruby, center=(620,992), url='https://lantern.dev/docs/languages/ruby'
StaticText , and
[802] link Rust, center=(693,992), url='https://lantern.dev/docs/languages/rust'
StaticText .
paragraph
StaticText Create a table with an embedding column
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText CREATE
StaticText TABLE
StaticText books
StaticText (id
StaticText SERIAL
StaticText PRIMARY KEY
StaticText , book_embedding
StaticText REAL
StaticText [3]);
paragraph
StaticText Generate embeddings
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText SELECT
StaticText text_embedding(
StaticText 'BAAI/bge-base-en'
StaticText ,
StaticText 'My text input'
StaticText );
StaticText 2
StaticText SELECT
StaticText openai_embedding(
StaticText 'openai/text-embedding-ada-002'
StaticText ,
StaticText 'My text input'
StaticText );
paragraph
StaticText Insert embeddings
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText INSERT INTO
StaticText books (book_embedding)
StaticText VALUES
StaticText (
StaticText '{0,1,0}'
StaticText ), (
StaticText '{3,2,4}'
StaticText );
paragraph
StaticText Calculate distance and select nearest rows
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText SELECT
StaticText book_embedding
StaticText <->
StaticText '{0,0,0}'
StaticText FROM
StaticText books
StaticText 2
StaticText ORDER BY
StaticText book_embedding
StaticText <->
StaticText '{0,0,0}'
StaticText LIMIT
StaticText 1
StaticText ;
paragraph
StaticText Create an index to speed up nearest neighbor search
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText CREATE
StaticText INDEX
StaticText book_index
StaticText ON
StaticText books
StaticText USING
StaticText lantern_hnsw(book_embedding dist_l2sq_ops)
StaticText 2
StaticText WITH
StaticText (M
StaticText =
StaticText 2
StaticText , ef_construction
StaticText =
StaticText 10
StaticText , ef
StaticText =
StaticText 4
StaticText , dim
StaticText =
StaticText 3
StaticText );
link Edit this page, url='https://github.com/lanterndata/docs/blob/main/010-develop/000-get-started.md'
paragraph
paragraph
StaticText ON THIS PAGE
list
[931] listitem, center=(1776,122), inner_text=Lantern Cloud
StaticText Lantern Cloud
[932] listitem, center=(1776,158), inner_text=Ubicloud
StaticText Ubicloud
[933] listitem, center=(1776,194), inner_text=Self-Host
StaticText Self-Host
[934] listitem, center=(1776,230), inner_text=Overview
StaticText Overview
RootWebArea Develop - Generate Embeddings | Lantern Docs, focused, url='https://lantern.dev/docs/develop/generate'
alert, atomic
StaticText Develop - Generate Embeddings | Lantern Docs
navigation
[567] link, center=(128,29), url='https://lantern.dev/'
Show more
image
[570] link Pricing, center=(786,29), url='https://lantern.dev/pricing'
[571] link Docs, center=(859,29), url='https://lantern.dev/docs'
[572] link Tutorials, center=(936,29), url='https://lantern.dev/tutorials'
[573] link Blog, center=(1012,29), url='https://lantern.dev/blog'
[574] link About, center=(1080,29), url='https://lantern.dev/about'
[575] link, center=(1144,29), url='https://github.com/lanterndata/lanterndb'
image
[578] link Login, center=(1742,29), url='https://lantern.dev/auth/sign-in'
[579] link Sign Up, center=(1842,29), url='https://lantern.dev/auth/sign-up'
main
navigation
[590] button Search K, center=(144,89), inner_text=Search
K
image
paragraph
image
paragraph
StaticText Develop
[598] link Get Started, center=(148,192), url='https://lantern.dev/docs/develop/get-started'
[600] link Store Embeddings, center=(148,224), url='https://lantern.dev/docs/develop/store'
[602] link Generate Embeddings, center=(148,256), focused, url='https://lantern.dev/docs/develop/generate'
[604] link Automatic Embedding Generation, center=(148,298), url='https://lantern.dev/docs/develop/columns'
[606] link Calculate Distance, center=(148,340), url='https://lantern.dev/docs/develop/distance'
[608] link Query Embeddings, center=(148,372), url='https://lantern.dev/docs/develop/query'
[610] link Create Index, center=(148,404), url='https://lantern.dev/docs/develop/indexing'
[612] link Quantization, center=(148,436), url='https://lantern.dev/docs/develop/quantization'
[614] link Asynchronous Tasks, center=(148,468), url='https://lantern.dev/docs/develop/async-tasks'
[616] link Weighted Vector Search, center=(148,500), url='https://lantern.dev/docs/develop/weighted-vector-search'
[618] link Troubleshooting, center=(148,532), url='https://lantern.dev/docs/develop/troubleshooting'
[620] link Single Operator, center=(148,564), url='https://lantern.dev/docs/develop/operator'
[622] link Postgres Notes, center=(148,596), url='https://lantern.dev/docs/develop/postgres'
[624] link Security, center=(148,628), url='https://lantern.dev/docs/develop/security'
paragraph
StaticText Languages
[628] link Javascript, center=(148,712), url='https://lantern.dev/docs/languages/javascript'
[630] link Python, center=(148,744), url='https://lantern.dev/docs/languages/python'
[632] link Ruby, center=(148,776), url='https://lantern.dev/docs/languages/ruby'
[634] link Rust, center=(148,808), url='https://lantern.dev/docs/languages/rust'
paragraph
StaticText Migrate
[638] link Migrate from Postgres to Lantern Cloud, center=(148,902), url='https://lantern.dev/docs/migrate/postgres-cloud'
[640] link Migrate from pgvector to Lantern Cloud, center=(148,954), url='https://lantern.dev/docs/migrate/pgvector-cloud'
[642] link Migrate from pgvector to self-hosted Lantern, center=(148,1006), url='https://lantern.dev/docs/migrate/pgvector-self'
[644] link Migrate from Pinecone to Lantern Cloud, center=(148,1058), url='https://lantern.dev/docs/migrate/pinecone'
paragraph
StaticText Lantern HNSW
link Installation, url='https://lantern.dev/docs/lantern-hnsw/install'
paragraph
StaticText Lantern Extras
link Installation, url='https://lantern.dev/docs/lantern-extras/install'
link Generate Embeddings, url='https://lantern.dev/docs/lantern-extras/embeddings'
link Lantern Daemon, url='https://lantern.dev/docs/lantern-extras/daemon'
paragraph
StaticText Lantern CLI
link Installation, url='https://lantern.dev/docs/lantern-cli/install'
link Generate Embeddings, url='https://lantern.dev/docs/lantern-cli/embeddings'
link Indexing Server, url='https://lantern.dev/docs/lantern-cli/indexing'
link Daemon, url='https://lantern.dev/docs/lantern-cli/daemon'
link Autotune Index, url='https://lantern.dev/docs/lantern-cli/autotune'
link Product Quantization, url='https://lantern.dev/docs/lantern-cli/pq'
paragraph
StaticText Contributing
link Dockerfile, url='https://lantern.dev/docs/contributing/docker'
link Visual Studio Code, url='https://lantern.dev/docs/contributing/vscode'
article
paragraph
StaticText Develop
heading Generate Embeddings
paragraph
StaticText Lantern supports generating text and image embeddings inside the database. Try it out on
[962] link Lantern Cloud, center=(1136,216), url='https://lantern.dev/'
StaticText .
paragraph
StaticText Note that generating embeddings is a compute-intensive task. For large scale embedding generation, such as generating embeddings over all of your data, Lantern provides a
[964] link separate process, center=(728,284), url='https://lantern.dev/docs/develop/generate'
StaticText .
heading Open AI Text Embeddings
paragraph
StaticText Before using Open AI text embeddings, you need to have an Open AI API key. You can get one by signing up at
[967] link Open AI, center=(1268,408), url='https://openai.com/'
StaticText . Once you have an API key, set it as a parameter in Postgres.
StaticText sql
[972] image, center=(1409,478)
StaticText Copy
code
StaticText 1
StaticText ALTER
StaticText ROLE
StaticText [YOUR_USERNAME]
StaticText SET
StaticText lantern_extras
StaticText .
StaticText openai_token
StaticText =
StaticText '[YOUR_API_KEY]'
StaticText ;
StaticText 2
StaticText SELECT
StaticText pg_reload_conf();
[973] span, center=(1436,478), inner_text=Copy
paragraph
StaticText Use the
code
StaticText openai_embedding
StaticText function to generate text embeddings using the Open AI embedding models. This function accepts a model name and text input as arguments, and for the
code
StaticText text-embedding-3-small
StaticText and
code
StaticText text-embedding-3-large
StaticText models, an optional dimension argument.
StaticText sql
[999] image, center=(1409,686)
StaticText Copy
code
StaticText 1
StaticText SELECT
StaticText openai_embedding(
StaticText 'openai/text-embedding-ada-002'
StaticText ,
StaticText 'My text input'
StaticText );
StaticText 2
StaticText SELECT
StaticText openai_embedding(
StaticText 'openai/text-embedding-3-large'
StaticText ,
StaticText 'My text input'
StaticText );
StaticText 3
StaticText SELECT
StaticText openai_embedding(
StaticText 'openai/text-embedding-3-large'
StaticText ,
StaticText 'My text input'
StaticText ,
StaticText 256
StaticText );
[1000] span, center=(1436,686), inner_text=Copy
paragraph
StaticText The following embedding models are supported
table
rowgroup
row
columnheader Model Name
paragraph
columnheader Dimensions
paragraph
columnheader Max Tokens
paragraph
rowgroup
row
[1042] cell openai/text-embedding-ada-002, center=(734,911)
paragraph
code
StaticText openai/text-embedding-ada-002
[1046] cell 1536, center=(1124,911)
paragraph
[1049] cell 8192, center=(1350,911)
paragraph
row
[1053] cell openai/text-embedding-3-small, center=(734,952)
paragraph
code
StaticText openai/text-embedding-3-small
[1057] cell 512 - 1536, center=(1124,952)
paragraph
[1060] cell 8192, center=(1350,952)
paragraph
row
[1064] cell openai/text-embedding-3-large, center=(734,993)
paragraph
code
StaticText openai/text-embedding-3-large
[1068] cell 256 - 3072, center=(1124,993)
paragraph
[1071] cell 8192, center=(1350,993)
paragraph
heading Cohere Text Embeddings
paragraph
StaticText Before using Cohere text embeddings, you need to have a Cohere API key. You can get one by signing up at
link Cohere, url='https://cohere.ai/'
StaticText . Once you have an API key, set it as a parameter in Postgres.
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText ALTER
StaticText ROLE
StaticText [YOUR_USERNAME]
StaticText SET
StaticText lantern_extras
StaticText .
StaticText cohere_token
StaticText =
StaticText '[YOUR_API_KEY]'
StaticText ;
StaticText 2
StaticText SELECT
StaticText pg_reload_conf();
paragraph
StaticText Use the
code
StaticText cohere_embedding
StaticText function to generate text embeddings using the Cohere embedding models. This function accepts a model name and text input as arguments, and an optional input type argument with values
code
StaticText search_document
StaticText or
code
StaticText search_query
StaticText (default is
code
StaticText search_query
StaticText ).
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText SELECT
StaticText cohere_embedding(
StaticText 'cohere/embed-english-v3.0'
StaticText ,
StaticText 'My text input'
StaticText );
StaticText 2
StaticText SELECT
StaticText cohere_embedding(
StaticText 'cohere/embed-english-v3.0'
StaticText ,
StaticText 'My text input'
StaticText ,
StaticText 'search_document'
StaticText );
paragraph
StaticText The following embedding models are supported
table
rowgroup
row
columnheader Model Name
paragraph
columnheader Dimensions
paragraph
columnheader Max Tokens
paragraph
rowgroup
row
cell cohere/embed-english-v3.0
paragraph
code
StaticText cohere/embed-english-v3.0
cell 1024
paragraph
cell 512
paragraph
row
cell cohere/embed-multilingual-v3.0
paragraph
code
StaticText cohere/embed-multilingual-v3.0
cell 1024
paragraph
cell 512
paragraph
row
cell cohere/embed-english-v2.0
paragraph
code
StaticText cohere/embed-english-v2.0
cell 4096
paragraph
cell 512
paragraph
row
cell cohere/embed-english-light-v2.0
paragraph
code
StaticText cohere/embed-english-light-v2.0
cell 1024
paragraph
cell 512
paragraph
row
cell cohere/embed-multilingual-v2.0
paragraph
code
StaticText cohere/embed-multilingual-v2.0
cell 768
paragraph
cell 512
paragraph
row
cell cohere/embed-english-light-v3.0
paragraph
code
StaticText cohere/embed-english-light-v3.0
cell 384
paragraph
cell 512
paragraph
row
cell cohere/embed-multilingual-light-v3.0
paragraph
code
StaticText cohere/embed-multilingual-light-v3.0
cell 384
paragraph
cell 512
paragraph
heading Open-Source Text Embeddings
paragraph
StaticText For example, to generate an embedding for the text
code
StaticText My text input
StaticText using the open-source embedding model
code
StaticText BAAI/bge-small-en
StaticText in SQL, run
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText SELECT
StaticText text_embedding(
StaticText 'BAAI/bge-small-en'
StaticText ,
StaticText 'My text input'
StaticText );
paragraph
StaticText The following embedding models are supported
table
rowgroup
row
columnheader Model Name
paragraph
columnheader Dimensions
paragraph
columnheader Max Tokens
paragraph
rowgroup
row
cell clip/ViT-B-32-textual
paragraph
code
StaticText clip/ViT-B-32-textual
cell 512
paragraph
cell 77
paragraph
row
cell microsoft/all-mpnet-base-v2
paragraph
code
StaticText microsoft/all-mpnet-base-v2
cell 768
paragraph
cell 128
paragraph
row
cell microsoft/all-MiniLM-L12-v2
paragraph
code
StaticText microsoft/all-MiniLM-L12-v2
cell 384
paragraph
cell 128
paragraph
row
cell transformers/multi-qa-mpnet-base-dot-v1
paragraph
code
StaticText transformers/multi-qa-mpnet-base-dot-v1
cell 768
paragraph
cell 250
paragraph
row
cell thenlper/gte-base
paragraph
code
StaticText thenlper/gte-base
cell 768
paragraph
cell 128
paragraph
row
cell thenlper/gte-large
paragraph
code
StaticText thenlper/gte-large
cell 1024
paragraph
cell 128
paragraph
row
cell llmrails/ember-v1
paragraph
code
StaticText llmrails/ember-v1
cell 1024
paragraph
cell 512
paragraph
row
cell intfloat/e5-base-v2
paragraph
code
StaticText intfloat/e5-base-v2
cell 768
paragraph
cell 512
paragraph
row
cell intfloat/e5-large-v2
paragraph
code
StaticText intfloat/e5-large-v2
cell 1024
paragraph
cell 512
paragraph
row
cell BAAI/bge-small-en
paragraph
code
StaticText BAAI/bge-small-en
cell 384
paragraph
cell 512
paragraph
row
cell BAAI/bge-base-en
paragraph
code
StaticText BAAI/bge-base-en
cell 768
paragraph
cell 512
paragraph
row
cell BAAI/bge-large-en
paragraph
code
StaticText BAAI/bge-large-en
cell 1024
paragraph
cell 512
paragraph
row
cell BAAI/bge-m3
paragraph
code
StaticText BAAI/bge-m3
cell 1024
paragraph
cell 8192
paragraph
row
cell jinaai/jina-embeddings-v2-small-en
paragraph
code
StaticText jinaai/jina-embeddings-v2-small-en
cell 512
paragraph
cell 8192
paragraph
row
cell jinaai/jina-embeddings-v2-base-en
paragraph
code
StaticText jinaai/jina-embeddings-v2-base-en
cell 768
paragraph
cell 8192
paragraph
heading Image Embeddings
paragraph
StaticText To generate image embeddings, use the
code
StaticText image_embedding
StaticText function. This function accepts a model name and image URL as arguments.
paragraph
StaticText For example, to generate an embedding for the image
code
StaticText https://lantern.dev/images/home/footer.png
StaticText using the embedding model
code
StaticText clip/ViT-B-32-visual
StaticText , run
StaticText sql
image
StaticText Copy
code
StaticText 1
StaticText SELECT
StaticText image_embedding(
StaticText 'clip/ViT-B-32-visual'
StaticText ,
StaticText 'https://lantern.dev/images/home/footer.png'
StaticText );
paragraph
StaticText The following embedding models are supported
table
rowgroup
row
columnheader Model Name
paragraph
columnheader Dimensions
paragraph
columnheader Max Tokens
paragraph
rowgroup
row
cell clip/ViT-B-32-visual
paragraph
code
StaticText clip/ViT-B-32-visual
cell 512
paragraph
cell 224
paragraph
heading Self-Hosting
paragraph
StaticText For people self-hosting, generating embeddings requires the Lantern Extras extension. Installation steps are found
link here, url='https://lantern.dev/docs/lantern-extras/install'
StaticText .
paragraph
StaticText Once the extension is installed, the above functions are available.
link Edit this page, url='https://github.com/lanterndata/docs/blob/main/010-develop/020-generate.md'
paragraph
paragraph
StaticText ON THIS PAGE
list
[1478] listitem, center=(1776,122), inner_text=Open AI Text Embeddings
StaticText Open AI Text Embeddings
[1479] listitem, center=(1776,158), inner_text=Cohere Text Embeddings
StaticText Cohere Text Embeddings
[1480] listitem, center=(1776,194), inner_text=Open-Source Text Embeddings
StaticText Open-Source Text Embeddings
[1481] listitem, center=(1776,230), inner_text=Image Embeddings
StaticText Image Embeddings
[1482] listitem, center=(1776,266), inner_text=Self-Hosting
StaticText Self-Hosting
Reasoning for action: By clicking on the 'Generate Embeddings' link, I am directly accessing the documentation section that aligns with my goal of understanding how to implement embedding generation in Lantern. This action is a crucial step in the process as it brings me closer to the acceptance criteria of my user story, which is to learn how to generate embeddings effectively for integration into my applications.
Action: click(602) # Click on the 'Generate Embeddings' link to explore how to generate embeddings with Lantern.
Change Description: The browser navigated from the 'Get Started' page to the 'Generate Embeddings' page of the Lantern Docs, changing the focused content and the displayed URL to reflect the new section regarding how to generate embeddings with Lantern.
Screenshot before:Screenshot after:---